Galileo (galileo.ai)
Treats memory as first-class in multi-agent tracing. Luna-2 SLMs (3B / 8B) scan every interaction for intent drift and belief drift; 20+ checks at sub-200ms latency. Catches when agent A's view of the world splits from teammate B's. OpenTelemetry-compatible.
At a glance
- Type
- Real-time intent / belief drift detection
- Tier
- T1
- Created
- 2021 (founded 2021 by Atindriyo Sanyal Vikram Chatterji Yash Sheth; seed May 2022)
- Latest release
- not applicable — not OSS
- License
- not applicable — not OSS
- GitHub
- not applicable — no GitHub repo
- Pricing
- free: 5K traces/mo; Pro: $100+/mo (50K traces); Enterprise: custom (VPC on-prem); runtime guardrails on Enterprise only
- Funding
- $68M total ($10M seed + $13M Series A + $45M Series B Oct 2024); Scale Venture Partners + Premji Invest led Series B
Taxonomy
- storage
- vector
- retrieval
- similarity
- persistence
- cross-session
- update
- append-only
- unit
- episode
- governance
- auditable
- conflict
- n/a
When to use
Optimised for: memory operation tracing + drift / poisoning detection
Anti-fit: not for use cases that don't run agent workloads in production
Pros & cons
Pros
First-mover in LLM observability; covers retrieval drift, hallucination, factuality alongside generic latency / cost — purpose-built for memory-driven agents.
Cons
Closed SaaS; pricing scales with traces; less open than Langfuse.
Claims & capabilities
Free 5k traces/month; Pro from $100; Enterprise custom. Doesn't scan stored memory files directly.
Technical surface
- API surface
- REST, SDK: Python
- Backend storage
- custom
- Deployment
- SaaS cloud; Enterprise: VPC or on-prem
- Embedding model
- locked
- Multi-tenancy
- hard-isolation
- MCP
- Yes — Galileo MCP Server documented; integrates with traces, prompts, datasets
- A2A
- Yes — A2A is listed under Integrations Overview (alongside CrewAI, Google ADK, LangChain, etc.)
- OpenTelemetry
- first-class — OTel-native ingestion
Compare Galileo (galileo.ai) with…
Similar systems
Other memory observability & monitoring in the catalog, ranked by inbound references.
- AgentOps T2
When Mem0 is connected, gains Memory Operation Timeline, Search Analytics, Memory Growth tracking, Error Tracking per memory call. Standalone, records context at each step but doesn't analyse memory quality.
- Langfuse T1
Memory module reads/writes captured as named spans. Trace Log View concatenates every agent step including memory ops. Agent Graphs (GA 2025) infer graph structure from observation nesting; session-level replay tracks how memory state evolves.
- LangSmith T1
Memory reads, vector DB retrievals, state changes are distinct span types in traces. RAG eval separates retrieval quality (context precision) from generation quality (faithfulness). Dataset versioning guards against eval drift.
- Ratine T2
Only tool found that scans the persistent memory layer on disk rather than runtime tracing. Detects injected instructions, obfuscated payloads (zero-width Unicode, base64, homoglyphs, hex), C2-pattern URLs, credential leakage. ratine diff compares snapshots for belief drift.